Br. J. exp. Path (I986) 67, 1-12 Pathogenesis of myonecrosis induced by coral snake (Micrurus nigrocinctus) venom in mice Jose Maria Gutierrez, Olga Arroyo, Fernando Chaves, Bruno Lomonte and Luis Cerdas Instituto Clodomiro Picado, Facultad de Microbiologia Universidad de Costa Rica, San Jose, Costa Rica, America Central Received for publication i i December I984 Accepted for publication 2I July I985 Summary. The mode by which coral snake (Micrurus nigrocinctus) venom affects skeletal muscle was studied using a combined approach. The venom induced early functional and structural alterations in the plasma membrane of muscle cells, suggesting that sarcolemma is the primary site of action of this venom. This was shown by the presence of wedge-shaped ('delta') lesions at the periphery of the cells, as well as by focal disruptions in the continuity of plasma membrane as early as 1 5 min after envenomation. After this initial alteration the rest of the organelles were severely affected. Myofilaments were hypercontracted leaving, as a consequence, areas of overstretched myofibrils as well as empty spaces. Eventually, myofilaments formed dense, clumped masses in which the striated structure was totally lost. At 24 h, myofilaments were still disorganized but they presented a more hyaline and homogeneous appearance. As early as 1 5 and 30 min mitochondria were swollen; later, by i, 3 and 24 h, they showed further alterations such as the presence of dense intracristal spaces and vesiculated cristae, as well as disruption in the integrity of their membranes. Sarcoplasmic reticulum was dilated and disorganized into many small vesicles randomly distributed throughout the cellular space. Moreover, the venom induced a rapid decrease in muscle levels of creatine and creatine-kinase (CK) and a calcium influx. Since the rates of efflux of creatine and CK were similar, it is suggested that the lesions produced in the membrane are large enough to allow the escape of these two molecules. As corroboration of the severe myotoxic effect, envenomated mice excreted reddish urine containing large quantities of myoglobin. Skeletal muscle cells are more susceptible to the action of the venom than erythrocytes, since coral snake venom induced only a mild direct haemolytic effect in vitro and haemolysis is not a significant effect in vivo. M. nigrocinctus venom induced a drastic increase in plasma levels of lactate dehydrogenase. Isozymes LDH-3, LDH-4, and LDH-5 increased markedly, suggesting that the systemic pathology of coral snake envenoming may be more complex than previously thought. Keywords: myonecrosis, coral snake venom, Micrurus nigrocinctus, myoglobinuria, haemolysis Correspondence: Dr Jose Maria Gutierrez, Instituto Clodomiro Picado, Facultad, de Microbiologia, Universidad de Costa Rica, San Jose, Costa Rica, Central America. J.M. Gutierrez et al. Many snake venoms induce skeletal muscle necrosis in humans and laboratory animals (Homma & Tu 197I; Gutierrez & Chaves I980; Mebs et al. I983). It has been demon- strated that some elapid venoms contain potent myotoxins, such as notexin and tai- poxin, isolated from the venoms of the Australian elapids Notechis scutatus and Oxyranus scutellatus, respectively (Harris et al. I975; Harris & Maltin I982). Studies on the pathogenesis of myonecrosis induced by snake venoms are important, not only to obtain a better understanding of envenoma- tions, but also to use these venoms as tools to approach more general aspects of muscle pathology. Coral snakes are the American representa- tives of the family Elapidae; they comprise more than 50 species (Roze I970). It has been known that coral snake venoms induce neurotoxic and cardiovascular effects (Jim&e nez-Porras et al. 1973; Brazil et al. I978). Recently, Gutierrez et al. (I983) demon- strated that they also exert a potent myotoxic activity in mice, suggesting the possibility that myotoxicity might be relevant in enve- nomings after coral snake bites. The present study was designed to obtain a better understanding of the pathogenesis of myonecrosis induced after experimental inoculations of venom of the Central Ameri- can coral snake Micrurus nigrocinctus. Par- ticular attention was given to the detection of early pathological changes in skeletal muscle cells. Materials and methods Venom. The venom used was a pool obtained from more than 6o specimens of Micrurus nigrocinctus collected in Costa Rica. Histology and ultrastructure. Groups of four mice (20-24 g) were injected im in the right thigh with venom (30 ug) in o.i ml of phosphate-buffered saline solution (pH 7.2). At several time intervals after injection (I 5 min, 30 min, i h, 3 h and 24 h) muscle samples were obtained from the enveno- mated thigh, cut into small pieces, and processed for light and electron microscopy as previously described (Arroyo & Gutierrez I98I). Thick sections were stained with toluidine blue and thin sections with uranyl acetate and lead citrate; the latter were examined in an Hitachi HU-I2A electron microscope. Control mice were injected with 0. I ml of saline solution. Changes in wet weight. Groups of four mice (20-24 g) were injected im in the right gastrocnemius with venom (30 Mg). At dif- ferent time intervals (30 min, i h and 3 h) mice were killed and both gastrocnemius muscles were obtained and weighed. Changes in wet weight of envenomated muscle were expressed as percentages, tak- ing as I00% the weight of the contralateral gastrocnemius. Changes in creatine kinase (CK) and creatine contents ofskeletal muscle. Groups of four mice (20-40 g) were injected im in the right gastrocnemius with venom (30 ,g). At several time intervals (30 min, i h and 3 h) they were killed and both gastrocnemius muscles were obtained. They were immedia- tely homogenized in 5 ml of phosphate- buffered saline containing O.I% Triton X- I00. Homogenates were then centrifuged at gooo g and the supernatants collected. An aliquot was diluted 35 times with distilled water and the CK content was estimated using the Sigma Kit Number 520 (Sigma Chemical Co., St Louis, MO, USA). The creatine content of another aliquot was determined (Sigma Kit Number 520). CK and creatine contents were expressed as percent- ages, taking as I00% the values of the controlateral gastrocnemius. Changes in calcium levels of muscle. Groups of four mice (20-24 g) were injected im in the right thigh with venom (30 ,ug). One and 3 h later, animals were killed and changes in cytoplasmic calcium levels of muscle fibres determined using the metallochromic cal- cium indicator Arsenazo III, according to 2 Myonecrosis by coral snake venom Gutierrez et al. (I984). Approximately 50 mg of muscle were transferred to a vial contain- ing: Arsenazo III o.i mm, digitonin 200 jug/ ml, NaCl 140 mM, KCL 5 mM, and imidazole IO mM (pH 7.0) in a total volume of 2.0 ml. After exactly i min, I.O ml of the superna- tant was transferred to a separate vial. The absorbance of the solution at 675 and 685 nm was read in a Varian-Techtron spectro- photometer. Calcium levels were directly proportional to the differential absorbance at 675-685 nm. As controls, groups of four mice were injected with O.I ml of saline solution; at i and 3 h pieces of muscle were obtained and calcium levels determined as described above. Detection of myoglobin in plasma and urine. Groups of four mice (20-24 g) were injected im with venom (30 jug). One and 3 h later, they were killed and samples of urine and plasma obtained and analysed electrophore- tically in I% agar (dissolved in Tris-EDTA- borate buffer, o.o6 M, pH 9.I). Electrophor- eses were performed using barbital buffer, 0.025 M, pH 8.6, according to Neremberg (I975). Standards of myoglobin and hae- moglobin were run in each determination. Haeme-containing proteins were detected by adding a colour reagent made of benzidine and hydrogen peroxide. Under these experi- mental conditions, myoglobin and haemog- lobin can be separated due to their different electrophoretic mobilities. Changes in lactate dehydrogenase (LDH) iso- zymes. Groups of four mice (20-24 g) were injected im in the thigh with venom (30 ug). At I, 3 and 6 h blood samples were collected from the tail into heparinized capillary tubes, and electrophoresis ofplasma was performed in order to detect changes in LDH isozymes, utilizing the Sigma Kit No. 705-EP (Sigma Chemical Co., St Louis, MO, USA). Myotoxic activity in vitro. The extensor digi- torum longus muscle of guinea pigs was dissected out, attached to a capillary tube, and was incubated at 3 70C in a test tube containing 5 ml of physiological salt solution bubbled with 95% °2: 5% CO2. The solution contained coral snake venom (50 ,ug/ml). At 30 min, i h, and 2 h samples of the solution were obtained and concentrations of CK and creatine were determined using Sigma Kit No. 520 (Sigma Chemical Co., St Louis, MO, USA). Control experiments, from which venom was excluded, were incubated and sampled at the same time intervals; control values were subtracted from values obtained in experiments with venom. Release of CK and creatine were expressed as percentages, taking as IOO% the release of these mole- cules in muscle incubated for 2 h with physiological salt solution containing i% Triton X-ioo. Experiments were repeated from four to six times. Direct haemolysis in vitro. Different concen- trations of coral snake venom were incu- bated with mouse erythrocytes previously washed five times with saline solution and diluted to a concentration of 2.5%. Tubes were incubated for i h at 3 70C and centri- fuged; absorbance of the supernatant was recorded at 54I nm, and haemolysis was expressed as a percentage, taking as IOO% the absorbance of the supernatant of eryth- rocytes incubated with distilled water. In order to test the influence ofcalcium in direct haemolysis, some experiments were per- formed with calcium (2 mM) whereas in others the assay mixture contained EDTA (2 mM) and no calcium. Results Light microscopy Muscles from animals injected with saline were histologically normal. In venom- injected mice, skeletal muscle damage was observed as early as I5 min after envenom- ing. The first alterations in muscle fibres were characterized by focal, wedge-shaped areas of degeneration at the periphery of the cells (Fig. i). These alterations are very similar to 'delta lesions' described in other muscle 3 4.M. Gutierrez et al. Fig. i. Photomicrograph of skeletal muscle taken 15 min after venom injection. Peripheral, wedge- shaped lesions ('delta lesions') are observed (arrows) in some cells, whereas others show a more advanced degeneration, with clumped masses of myofilaments (M). x 360. *40J Fig. 2. Electron micrograph of skeletal muscle I 5 min after injection ofM. nigrocinctus venom. Portion ofa cell with interruptions in the integrity of the sarcolemma (arrows). The myofibrils have not yet been affected. Notice the presence of an intact basal lamina. x 35 000. 4 ......... Myonecrosis by coral snake venom pathologies such as Duchenne muscular dystrophy (eg Mokri & Engel I975). At I5 min, a few cells were in a more advanced stage of degeneration and their cytoplasm was disorganized into amorphous, dense clumps. By i h and 3 h there was widespread muscle damage in which cells were drasti- cally affected, with clumped myofibrils alter- nating with empty spaces in the cytoplasm. At 24 h necrotic fibres had a more hyaline appearance and distribution of myofibrillar material was more homogeneous; at this period, necrotic cells had been invaded by phagocytes. M. nigroinctus venom did not induce vascular alterations such as haemor- rhage and thrombosis. Electron microscopy The ultrastructure of control muscle was normal in all samples examined. On the other hand, in venom-injected muscle the plasma membrane was damaged soon after envenoming. At I 5 min many cells had focal interruptions in the integrity of the sarco- lemma (Fig. 2). Since some cells had alte- rations only at the plasma membrane; this indicates that the sarcolemma was altered before any other organelle. Also at I5 min many cells showed peripheral areas ofdegen- eration in which the plasma membrane was disrupted or totally absent. Beneath these focal sarcolemmal alterations there were localized areas of disorganization of myofi- brils (Fig. 3). In these areas, the basal lamina remained intact at the periphery of the cell (Fig. 3). The first alterations observed in myofilaments, at 1 5 min, were characterized by areas of hypercontraction alternating with areas of overstretched myofibrils (Fig. 4). In some cells, the areas of hypercontrac- tion alternated with empty spaces in the cytoplasm. By 30 min myofibrillar material was disorganized into amorphous, dense clumps in which the structure was totally lost. At 30 min, i h and 3 h many skeletal muscle fibres were irreversibly injured. Their Fig. 3. Electron micrograph of skeletal muscle taken I 5 min after injection of M. nigrocinctus venom. Peripheral area of a necrotic fibre in which the sarcolemma is absent although the basal lamina is intact (BL). Myofilaments (My) are disorganized into a dense clump and mitochondria (Mi) are swollen. x 5 200. 5 J.M. Gutirrez et al. Fig. 4. Electron micrograph of skeletal muscle. Changes in myofibrils I 5 min after envenomation. Early alterations in which areas of hypercontraction alternate with overstretched portions. This stage precedes the formation of clumped masses of myofilaments. x 4500 ------- ----- Fig. 5. Electron micrograph of skeletal muscle 24 h after injection of M. nigrocinctus. Myofibrillar material is disorganized to form a hyaline, homogeneous mass of myofilaments in which the normal striated pattern has been lost. Sarcoplasmic reticulum (arrow) shows moderate dilation. x 5200. 6 Myonecrosis by coral snake venom plasma membranes were missing although the basal lamina was apparently intact. Myofilaments were clumped into dense masses that left many empty spaces in the cytoplasm. In some areas, myofibrillar-like material was observed in the interstitial space. Twenty-four hours after envenoming the disposition of the myofilaments had changed to a less dense, more hyaline pat- tern with a more homogeneous distribution in the cellular space (Fig. 5). Sarcoplasmic reticulum was severely affected by 15 min after envenoming, and there were many small vesicles distributed throughout the cellular space. Furthermore, T-tubules were not observed in necrotic muscle, indicating that they also were disrupted. Mitochondria showed sequential damage; by I5 and 30 min they were swollen but many of them retained their internal structure, but after longer time periods (i h, 3 h and 24 h) they were severely damaged. Many were swollen with dense intracristal spaces and vesicu- lated cristae; some contained only one mem- brane and were disrupted (Fig. 6). Nuclei also were severely affected with dense chromatin and disruption of the nuclear envelope in many cells. In samples obtained 24 h after envenoming there was an abundant phago- cytic cell infiltrate in the interstitial space and inside necrotic muscle fibres. M. nigro- cinctus venom did not affect the morphology of capillary vessels. Changes in wet weight, CK, creatine and calcium There was a moderate increase in wet weight of envenomated gastrocnemius, reaching a value of I12+3% (n=4) at 3 h (Fig. 7). A parallel decrease in muscle creatine and CK was observed (Fig. 7) from as early as 30 min, thus reflecting functional alterations in the plasma membrane. By 3 h, CK and creatine levels were 35.7±8% (n=4) and 34.I ± 2% (n = 4), respectively, by compari- son with controls. Moreover, a prominent ,$'4.' ;j g^ + tvs